One interesting feature of some synthetic diamonds is their attraction to a simple magnet. (Strictly speaking, diamonds do not possess any magnetism.) This apparent “magnetism” is due to the existence of metallic inclusions – usually the flux materials, such as Fe, Ni, or Co, used in the growth process. While magnetism is a very good indication that a diamond is synthetic, some rare exceptions do exist.

Recently, a 0.12 ct Fancy Intense purplish pink diamond submitted to GIA’s West Coast laboratory demonstrated apparent magnetism (see figure). In addition to pink graining, magnification revealed several inclusions, including “feathers” and dark features, a few of which reached the surface. Some of these features had a platy habit, while others were irregular in shape. The diamond also had several cavities on the girdle and pavilion facets that were filled with a dark material. It had a few cloud-like patches of pinpoint inclusions, but their appearance was different from the clouds we have seen in synthetic diamonds. When observed with a desk-model spectroscope, the diamond showed only a 415 nm “ Cape” line. It fluoresced blue to standard long-wave UV radiation, and yellow and blue to short-wave UV. When examined with the Diamond Trading Company (DTC) DiamondView™ fluorescence imaging system, it did not show the cross-shaped zoning typically seen in most synthetic diamonds. Last, no bands that could be attributed to Ni or other flux metals were observed in the photoluminescence spectrum. On the basis of these results, we concluded that the diamond was natural.

When a magnet was brought close to the diamond, however, it attracted the dark material that was exposed in cavities along the girdle. When we analyzed this dark material using energy-dispersive X-ray fluorescence, we found Fe and Ca to be the major components. We studied the exposed dark material further using Raman spectroscopy and found features that could be attributed to hematite, limonite (rust), and diamond. We concluded that this dark material was residue from the manufacturing process. Metallic fragments within the residue caused the magnetic behavior of this sample, while oxidation of the iron fragments generated the hematite and limonite peaks observed in the Raman spectra. We attempted to analyze the dark inclusions in the diamond with Raman spectroscopy, but the spectra did not reveal any recognizable features. Raman analyses of some of these inclusions where they reached the surface produced spectra similar to the spectrum of the dark material in the cavities, but this also may be from residue covering those inclusions.

The magnetic response of this natural diamond was similar in intensity to what we have seen in synthetic diamonds with metallic inclusions. Cleaning the diamond with acid would presumably dissolve this residue, and thereby render the diamond “nonmagnetic.” Attraction to a magnet remains a useful way to check for synthetic diamonds, but as with other gem testing methods, this should not be the sole identification criterion.

Andy H. Shen, GIA Gem Laboratory, Carlsbad
James E. Shigley, GIA Research, Carlsbad